A method and system for detecting vegetation change based on multi-temporal satellite optical remote sensing data

Abstract

This invention provides a method and system for detecting vegetation change based on multi-temporal satellite optical remote sensing data, comprising: acquiring two consecutive remote sensing images based on satellite observations; removing cloud-containing pixels from the two consecutive remote sensing images to obtain remote sensing data of the non-cloud area in the observation region; calculating the normalized vegetation index (NDI) of the non-cloud area using the channel reflectance of the non-cloud area; identifying vegetation and non-vegetation pixels in the non-cloud area to obtain a binary vegetation distribution map; and segmenting the binary vegetation distribution map into several sub-map regions to determine the vegetation change. This invention can be used in the field of satellite remote sensing observation of short-period local vegetation change research and is an effective and important technical means for monitoring ecological and environmental changes.

Description

Technical Field

[0001] This invention relates to the field of satellite remote sensing vegetation change detection technology, specifically to a method and system for detecting vegetation change based on multi-temporal satellite optical remote sensing data. Background Technology

[0002] Vegetation refers to all plants covering the Earth's surface and is a key area of ​​research in global carbon and water cycles, terrestrial-atmospheric radiation budgets, and related ecological fields, playing a crucial role in ecosystems. Vegetation phenological characteristics are often influenced by natural factors such as precipitation, sunlight, and topography. In addition, anthropogenic factors, such as deforestation, mining, and urbanization, as well as natural disasters, such as fires, mudslides, and earthquakes, also have a profound impact on vegetation growth. Therefore, vegetation change can largely reflect the safety and stability of ecosystems and is an important indicator of ecological and environmental changes.

[0003] Currently, common methods for detecting vegetation change include using long-term satellite remote sensing data to obtain long-term trends in vegetation change, or using ground observation stations for observation. However, neither of these methods can obtain detailed information on local changes in vegetation over a short period.

[0004] Patent document CN115205685A discloses an automated monitoring method for vegetation change in open-pit mining areas based on long-term time-series remote sensing data, belonging to the field of remote sensing and geographic information technology. The method includes the following steps: S1, Data acquisition and preprocessing: Acquire all data from the mining area during the study period, and obtain NDVI time-series data through cropping, NDVI calculation, maximum value synthesis, and filtering; S2, Template creation: Based on statistical analysis of the filtered NDVI time-series data, obtain the NDVI values ​​of vegetation and bare soil in the mining area, and combine this with typical vegetation change patterns in open-pit mining areas to generate NDVI templates for different vegetation change trajectory types; S3, Vegetation change trajectory type classification: Using the NDVI templates as training data, obtain the vegetation change trajectory type for each pixel in the mining area using a DTW-based KNN classification method.

[0005] However, patent document CN115205685A has limitations in monitoring vegetation changes caused by sudden events, namely, it cannot effectively detect short-term local vegetation changes. The method in that patent document requires the use of long-term remote sensing data to establish NDVI templates for different vegetation change trajectory types. In this process, the average characteristics of vegetation distribution are often obtained, but the vegetation change characteristics in a shorter period cannot be reflected. This method is suitable for monitoring long-term vegetation change trends, but its effectiveness in detecting vegetation feature changes caused by sudden natural or human events in a short period is limited. In contrast, the vegetation change detection method and system based on multi-temporal satellite optical remote sensing data introduced in this patent uses two consecutive multi-temporal remote sensing images of the same area as a basis. By comparing the changes in vegetation features detected by remote sensing before and after, the vegetation change characteristics are obtained. It does not require the accumulation of long-term remote sensing data to build complex models and is suitable for detecting local vegetation changes in a short period of time.

[0006] Therefore, it is necessary to propose a vegetation change detection method based on multi-temporal satellite optical remote sensing data for effective detection of vegetation changes. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for detecting vegetation changes based on multi-temporal satellite optical remote sensing data.

[0008] A vegetation change detection method based on multi-temporal satellite optical remote sensing data provided by the present invention includes:

[0009] Step S1: Obtain remote sensing data for both the preceding and following scenes based on satellite observations;

[0010] Step S2: Remove cloud-containing pixels from the two sets of remote sensing data to obtain remote sensing data of the non-cloud areas in the observation area;

[0011] Step S3: Calculate the normalized vegetation index of the non-cloudy area using the channel reflectance of the non-cloudy area;

[0012] Step S4: Distinguish between vegetation and non-vegetation pixels in non-cloud areas to obtain a binary map of vegetation distribution;

[0013] Step S5: Divide the binary map of vegetation distribution into several sub-map regions to determine the vegetation change.

[0014] Preferably, the satellite observation range is consistent, and the remote sensing data includes channel reflectance after positioning and calibration, as well as pixel location information.

[0015] Preferably, in step S2, removing cloud-covered pixels using a cloud mask involves judging each pixel within the observation area, using existing cloud mask data to remove pixels marked as cloud-covered, and obtaining the channel reflectance of the non-cloud area and the location information of non-cloud pixels.

[0016] Preferably, in step S3, the channel reflectance of the non-cloudy area includes the red channel reflectance and near-infrared channel reflectance of the non-cloudy pixels. Cloudy pixels are not processed. The normalized vegetation index of the non-cloudy area within the observation range is obtained, and the calculation formula is as follows:

[0017]

[0018] Wherein, NDVI represents the Normalized Difference Vegetation Index, R RED R represents the reflectance of the red channel. INR This indicates the reflectivity of the near-infrared channel.

[0019] Preferably, in step S4, when the normalized vegetation index of a pixel is greater than 0.1, it is determined to be vegetation and marked as 0;

[0020] When the normalized vegetation index of a pixel is less than or equal to 0.1, it is judged as non-vegetation and marked as 1.

[0021] Preferably, step S5 includes the following sub-steps:

[0022] Step S5.1: Divide the binary map of vegetation distribution according to a set range to obtain several sub-map regions;

[0023] Step S5.2: Calculate the distribution of non-vegetation cell connected components in each sub-map region of the vegetation distribution binary map;

[0024] Step S5.3: Compare the distribution characteristics of non-vegetation pixel connected regions in the corresponding sub-image regions of the two vegetation distribution binary images to determine the vegetation changes and the locations of areas with significant changes.

[0025] Preferably, the distribution features include the location of the largest connected region of non-vegetated pixels and the number of pixels it contains.

[0026] The determination of vegetation change and the location of areas with significant changes includes: comparing the center position of the largest non-vegetated pixel connected region and the change in the number of pixels in the corresponding sub-image regions of the two remote sensing images. If the change in both features exceeds 75%, then the center position of the connected region can represent the location of areas with significant vegetation changes, and the change in the number of pixels can represent the vegetation change.

[0027] A vegetation change detection system based on multi-temporal satellite optical remote sensing data, provided by the present invention, includes:

[0028] Module M1: Acquires remote sensing data from two consecutive scenes based on satellite observations;

[0029] Module M2: Removes cloud-containing pixels from the two sets of remote sensing data to obtain remote sensing data of the non-cloud areas in the observation area;

[0030] Module M3: Calculates the normalized vegetation index of the non-cloudy area using the channel reflectance of the non-cloudy area;

[0031] Module M4: Distinguish between vegetation and non-vegetation pixels in non-cloud areas to obtain a binary map of vegetation distribution;

[0032] Module M5: Performs regional segmentation on the binary map of vegetation distribution to obtain several sub-map regions, thereby determining the vegetation change situation.

[0033] According to the present invention, a computer-readable storage medium storing a computer program is provided, wherein when the computer program is executed by a processor, the steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data are implemented.

[0034] An electronic device according to the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the computer program, when executed by the processor, implements the steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. This invention detects vegetation changes based on multi-temporal satellite optical remote sensing data. Existing vegetation change detection methods mostly focus on long-term changes in vegetation phenological characteristics, but there is little technical support for short-term local vegetation changes caused by sudden natural or human factors. This invention fills the gap in the existing technology.

[0037] 2. This invention can be used in the field of satellite remote sensing observation of short-period local vegetation changes, and is an effective and important technical means for monitoring ecological and environmental changes.

[0038] 3. This invention proposes a vegetation change detection method based on multi-temporal satellite optical remote sensing data for short-term local vegetation changes. By identifying vegetation features in remote sensing images within the same observation range and comparing them before and after, local vegetation changes can be effectively detected, and the locations of the most significant changes can be identified. Furthermore, unlike methods that focus on the average characteristics of vegetation within the observation range, this method fully considers the complexity of local vegetation change characteristics by dividing the observation range into regions, thus effectively improving the detection efficiency of vegetation changes. Attached Figure Description

[0039] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0040] Figure 1 This is a schematic diagram of the working method of the present invention.

[0041] Figure 2 This is a schematic diagram of two optical remote sensing data of the same area observed by satellite in this invention.

[0042] Figure 3 These are two normalized vegetation index distribution maps of non-cloudy areas in the same region observed by satellite in this invention.

[0043] Figure 4 This is a distribution map of connected domains of two non-cloudy, non-vegetation pixel areas in the same region observed by satellite in this invention.

[0044] Figure 5 This is a schematic diagram showing the location of areas with significant vegetation changes in this invention. Detailed Implementation

[0045] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0046] This invention obtains multi-temporal optical remote sensing data of a certain area based on satellite observations, mainly including the reflectance of the observation channel and pixel location information; combined with cloud masking of the observation area, it determines the reflectance of the non-cloudy channel and the location information of non-cloudy pixels; based on the reflectance of the satellite observation channel, it calculates the normalized vegetation index of the non-cloudy area; using a threshold method, it identifies vegetation and non-vegetation pixels in the non-cloudy area and obtains a binary map of vegetation distribution; it divides the binary map of vegetation distribution into multiple sub-map regions and obtains the connected component distribution of non-vegetation pixels in the sub-map regions; it compares the changes in the connected components of non-vegetation pixels in the sub-map regions of the multi-temporal vegetation distribution binary map to obtain the vegetation changes and the locations of areas with significant changes within the satellite observation area.

[0047] The present invention provides a method for detecting vegetation change based on multi-temporal satellite optical remote sensing data, such as... Figure 1 As shown, it includes the following steps:

[0048] Step S1: Obtain remote sensing data for two consecutive scenes based on satellite observations. The satellite observation range is consistent, and the remote sensing data includes the channel reflectance and pixel location information after positioning and calibration.

[0049] Step S2: Remove cloud-covered pixels from the two sets of remote sensing data to obtain remote sensing data for the non-cloudy areas of the observation region. Specifically, remove cloud-covered pixels from the observation region to obtain the channel reflectance and non-cloudy pixel location information for the non-cloudy areas. Removing cloud-covered pixels using a cloud mask involves classifying each pixel within the observation region and using existing cloud mask data to remove pixels marked as having clouds, thus obtaining the channel reflectance and non-cloudy pixel location information for the non-cloudy areas.

[0050] Step S3: Calculate the normalized vegetation index (NDI) for the non-cloudy area using the channel reflectance of the non-cloudy area. The channel reflectance of the non-cloudy area includes the red channel reflectance and near-infrared channel reflectance of the non-cloudy pixels. Cloudy pixels are not processed. The NDI for the non-cloudy area within the observation range is obtained as follows:

[0051]

[0052] Wherein, NDVI represents the Normalized Difference Vegetation Index, R RED R represents the reflectance of the red channel. INR This indicates the reflectivity of the near-infrared channel.

[0053] Step S4: Distinguish between vegetation and non-vegetation pixels in non-cloudy areas to obtain a binary map of vegetation distribution. Distinguishing between vegetation and non-vegetation pixels in non-cloudy areas includes using a threshold method. An empirical value of 0.1 is selected as the threshold for distinguishing between vegetation and non-vegetation pixels. Pixels with cloud cover are left unprocessed. If the normalized vegetation index (NDI) of a pixel is greater than 0.1, it is classified as vegetation and marked as 0; if the NDI is less than or equal to 0.1, it is classified as non-vegetation and marked as 1. This process yields the binary map of vegetation distribution.

[0054] Step S5: Segment the binary map of vegetation distribution to obtain several sub-map regions, thereby determining the vegetation change. Step S5 includes the following sub-steps:

[0055] Step S5.1: Divide the binary map of vegetation distribution into several sub-map regions according to a set range. Dividing according to the set range includes selecting a 100*100 pixel range as the set range to divide the binary map of vegetation distribution into multiple sub-map regions.

[0056] Step S5.2: Calculate the connected component distribution of non-vegetation pixels in each sub-region of the binary vegetation distribution map. For each sub-region of the entire vegetation distribution binary map, the connected component distribution of non-vegetation pixels is calculated. That is, in each sub-region, pixels marked as 0 are vegetation pixels, and pixels marked as 1 are non-vegetation pixels. For each non-vegetation pixel in the sub-region, if its position is marked as (i, j), where i represents the row number and j represents the column number in the image, then the eight neighboring points around the current pixel, i.e., (i-1, j), (i+1, j), (i, j-1), (i, j+1), (i-1, j-1), (i-1, j+1), (i+1, j-1), (i+1, j+1), if there is a non-vegetation pixel among them, it is marked into the same connected component of pixel (i, j). Finally, the connected component distribution map of non-vegetation pixels in the sub-region is obtained.

[0057] Step S5.3: Compare the distribution features of non-vegetation pixel connected components in corresponding sub-image regions of the two vegetation distribution binary images to determine the vegetation change and the location of areas with significant changes. Compare the distribution features of non-vegetation pixel connected components in corresponding sub-image regions of the two remote sensing images. These features include the location of the largest connected component of non-vegetation pixels and the number of pixels it contains. Determining the vegetation change and the location of areas with significant changes involves comparing the center location of the largest non-vegetation pixel connected component and the change in the number of pixels in the corresponding sub-image regions of the two remote sensing images. If the change in both features exceeds 75%, then the center location of the connected component can represent the location of areas with significant vegetation change, and the change in the number of pixels can represent the vegetation change.

[0058] Furthermore, in conjunction with the appendix Figures 1 to 5 The vegetation change detection method based on multi-temporal satellite optical remote sensing data of the present invention is described in detail below:

[0059] The prerequisite for vegetation change detection is acquiring multi-temporal satellite remote sensing data. Based on satellite observations, two consecutive remote sensing images are obtained, with the satellite observation area being consistent. The remote sensing data mainly includes channel reflectance after positioning and calibration, as well as pixel location information. A schematic diagram of two optical remote sensing images of the same area observed by satellite is shown below. Figure 2 As shown.

[0060] By using cloud masking, all pixels in the observation area are removed from the cloud-covered areas. This involves judging each pixel individually and using existing cloud masking data to remove pixels marked as having clouds, thereby obtaining the channel reflectance of the non-cloudy areas and the location information of the non-cloudy pixels.

[0061] The normalized vegetation index (NVI) for the non-cloudy area is calculated using the red and near-infrared reflectance of non-cloudy pixels. Cloudy pixels are left unprocessed. This yields the NVI distribution map of the non-cloudy area within the observation range. Two satellite observation maps of the same region's non-cloudy NVI distribution are shown below. Figure 3 As shown:

[0062]

[0063] Wherein, NDVI represents the Normalized Difference Vegetation Index, R RED R represents the reflectivity of the red channel. INR This represents the reflectivity of the near-infrared channel.

[0064] Based on the Normalized Difference Vegetation Index (NDVI) of non-cloudy areas, a threshold method is used to distinguish between vegetation and non-vegetation pixels in non-cloudy areas. An empirical value of 0.1 is selected as the threshold for classifying non-cloudy pixels, while cloudy pixels are left unprocessed. Pixels with an NDVI greater than 0.1 are classified as vegetation and marked as 0, while pixels with an NDVI less than or equal to 0.1 are classified as non-vegetation and marked as 1, thus obtaining a binary map of vegetation distribution.

[0065] For the binary map of vegetation distribution, it is divided according to a set range. A 100*100 pixel range is selected as the set range to divide the binary map of vegetation distribution into multiple sub-map regions, resulting in several sub-map regions.

[0066] For each sub-image region in the binary image of the overall vegetation distribution, the connected component calculation of non-vegetation pixels is performed. That is, in each sub-image region, pixels marked 0 are vegetation pixels, and pixels marked 1 are non-vegetation pixels. For each non-vegetation pixel in the sub-image region, if its position is marked as (i, j), where i represents the row number and j represents the column number, then among the eight neighboring points around the current pixel—namely (i-1, j), (i+1, j), (i, j-1), (i, j+1), (i-1, j-1), (i-1, j+1), (i+1, j-1), (i+1, j+1)—if any of them contain non-vegetation pixels, they are marked into the same connected component as pixel (i, j). This ultimately yields the connected component distribution map of non-vegetation pixels in the sub-region. The connected component distribution maps of non-vegetation pixels in two non-cloudy areas of the same region observed by satellite are shown below. Figure 4 As shown in the figure, the color fill indicates the distribution of connected domains, and the pentagram indicates the center position of the connected domain of the largest non-vegetated cell in the sub-figure region.

[0067] Comparing the distribution characteristics of non-vegetation pixel connected regions in corresponding sub-regions of two binary vegetation distribution maps involves comparing the distribution characteristics of non-vegetation pixel connected regions in corresponding sub-regions of the two remote sensing images. These characteristics include the location of the largest non-vegetation pixel connected region and the number of pixels it contains. By comparing the changes in the center location and the number of pixels in the largest non-vegetation pixel connected region in corresponding sub-regions of the two remote sensing images, if the changes in both features exceed 75%, then the center location of this connected region can represent the location of an area with significant vegetation change, and the change in the number of pixels can represent the degree of vegetation change. Figure 5 The five-pointed star in the center indicates the location of areas with significant vegetation changes.

[0068] The present invention also provides a vegetation change detection system based on multi-temporal satellite optical remote sensing data. The vegetation change detection system based on multi-temporal satellite optical remote sensing data can be implemented by executing the process steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data. That is, those skilled in the art can understand the vegetation change detection method based on multi-temporal satellite optical remote sensing data as a preferred embodiment of the vegetation change detection system based on multi-temporal satellite optical remote sensing data.

[0069] A vegetation change detection system based on multi-temporal satellite optical remote sensing data, provided by the present invention, includes:

[0070] Module M1: Acquires remote sensing data from two consecutive satellite images. The satellite observation range is consistent, and the remote sensing data includes channel reflectance after positioning and calibration, as well as pixel location information.

[0071] Module M2: Removes cloud-covered pixels from the two consecutive remote sensing images to obtain remote sensing data for the non-cloudy areas of the observation region. Module M2 uses cloud masking to remove cloud-covered pixels, including pixel-by-pixel judgment within the observation region, using existing cloud masking data to remove pixels marked as having clouds, and obtaining the channel reflectance of the non-cloudy areas and the location information of the non-cloudy pixels.

[0072] Module M3: Calculates the Normalized Difference Vegetation Index (NDVI) for cloudless areas using channel reflectance. The channel reflectance for cloudless areas in Module M3 includes the red channel reflectance and near-infrared channel reflectance of cloudless pixels. Cloudy pixels are not processed. The NDVI for the observed cloudless area is obtained using the following formula:

[0073]

[0074] Wherein, NDVI represents the Normalized Difference Vegetation Index, R RED R represents the reflectance of the red channel. INR This indicates the reflectivity of the near-infrared channel.

[0075] Module M4: Distinguish between vegetation and non-vegetation pixels in non-cloud areas to obtain a binary map of vegetation distribution. In Module M4, pixels with a normalized vegetation index (NDI) greater than 0.1 are classified as vegetation and marked as 0; pixels with an NDI less than or equal to 0.1 are classified as non-vegetation and marked as 1.

[0076] Module M5: Segment the binary vegetation distribution map into several sub-map regions to determine vegetation changes. Module M5 includes the following sub-steps: Module M5.1: Segment the binary vegetation distribution map according to a set range to obtain several sub-map regions. Module M5.2: Calculate the non-vegetation pixel connected component distribution in each sub-map region of the binary vegetation distribution map. Module M5.3: Compare the non-vegetation pixel connected component distribution characteristics of corresponding sub-map regions in two images of the binary vegetation distribution map to determine vegetation changes and the location of areas with significant changes. Distribution characteristics include the location of the largest connected component of non-vegetation pixels and the number of pixels it contains. Determining vegetation changes and the location of areas with significant changes includes: comparing the center position and pixel number changes of the largest non-vegetation pixel connected component in the corresponding sub-map regions of the two remote sensing images. If the changes in both features exceed 75%, then the center position of the connected component can represent the location of areas with significant vegetation changes, and the change in the number of pixels can represent the vegetation changes.

[0077] According to the present invention, a computer-readable storage medium storing a computer program is provided, wherein when the computer program is executed by a processor, the steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data are implemented.

[0078] An electronic device according to the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the computer program, when executed by the processor, implements the steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data.

[0079] Those skilled in the art will understand that, besides implementing the system and its various devices, modules, and units provided by this invention in the form of purely computer-readable program code, the same functions can be achieved entirely through logical programming of the method steps, making the system and its various devices, modules, and units of this invention function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, the system and its various devices, modules, and units provided by this invention can be considered as a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; alternatively, the devices, modules, and units for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0080] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for detecting vegetation change based on multi-temporal satellite optical remote sensing data, characterized in that, include: Step S1: Obtain remote sensing data for both the preceding and following scenes based on satellite observations; Step S2: Remove cloud-containing pixels from the two sets of remote sensing data to obtain remote sensing data of the non-cloud areas in the observation area; Step S3: Calculate the normalized vegetation index of the non-cloudy area using the channel reflectance of the non-cloudy area; Step S4: Distinguish between vegetation and non-vegetation pixels in non-cloud areas to obtain a binary map of vegetation distribution; Step S5: Divide the binary map of vegetation distribution into several sub-map regions to determine the vegetation change. Step S5 includes the following sub-steps: Step S5.1: Divide the binary map of vegetation distribution according to a set range to obtain several sub-map regions; Step S5.2: Calculate the distribution of non-vegetation cell connected components in each sub-map region of the vegetation distribution binary map; Step S5.3: Compare the distribution characteristics of non-vegetation pixel connected regions in the corresponding sub-image regions of the two vegetation distribution binary images to determine the vegetation changes and the locations of areas with significant changes; The distribution features include the location of the largest connected region of non-vegetated pixels and the number of pixels it contains; The determination of vegetation change and the location of areas with significant changes includes: comparing the center position of the largest non-vegetated pixel connected region and the change in the number of pixels in the corresponding sub-image regions of the two remote sensing images. If the change in both features exceeds 75%, then the center position of the connected region can represent the location of areas with significant vegetation changes, and the change in the number of pixels can represent the vegetation change.

2. The vegetation change detection method based on multi-temporal satellite optical remote sensing data according to claim 1, characterized in that, The satellite observation range is consistent, and the remote sensing data includes channel reflectance and pixel location information after positioning and calibration.

3. The vegetation change detection method based on multi-temporal satellite optical remote sensing data according to claim 2, characterized in that, In step S2, cloud masking is used to remove cloud-covered pixels. This includes judging each pixel in the observation area, using existing cloud masking data to remove pixels marked as cloud-covered, and obtaining the channel reflectance of the non-cloud area and the location information of non-cloud pixels.

4. The vegetation change detection method based on multi-temporal satellite optical remote sensing data according to claim 2, characterized in that, In step S3, the channel reflectance of the non-cloudy area includes the red channel reflectance and near-infrared channel reflectance of the non-cloudy pixels. Cloudy pixels are not processed. The normalized vegetation index of the non-cloudy area within the observation range is obtained, and the calculation formula is as follows: Wherein, NDVI represents the Normalized Difference Vegetation Index, R RED R represents the reflectance of the red channel. INR This indicates the reflectivity of the near-infrared channel.

5. The vegetation change detection method based on multi-temporal satellite optical remote sensing data according to claim 1, characterized in that, In step S4, when the normalized vegetation index of a pixel is greater than 0.1, it is determined to be vegetation and marked as 0; When the normalized vegetation index of a pixel is less than or equal to 0.1, it is judged as non-vegetation and marked as 1.

6. A system for implementing the vegetation change detection method based on multi-temporal satellite optical remote sensing data as described in claim 1, characterized in that, include: Module M1: Acquires remote sensing data from two consecutive scenes based on satellite observations; Module M2: Removes cloud-containing pixels from the two sets of remote sensing data to obtain remote sensing data of the non-cloud areas in the observation area; Module M3: Calculates the normalized vegetation index of the non-cloudy area using the channel reflectance of the non-cloudy area; Module M4: Distinguish between vegetation and non-vegetation pixels in non-cloud areas to obtain a binary map of vegetation distribution; Module M5: Performs regional segmentation on the binary map of vegetation distribution to obtain several sub-map regions, thereby determining the vegetation change situation.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data as described in any one of claims 1 to 5.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the vegetation change detection method based on multi-temporal satellite optical remote sensing data as described in any one of claims 1 to 5.

Images